Comprehensive Study on the Thrust Estimation and Anti-Freezing Lubricant of Pipe Jacking in Frozen Soil

Abstract

Recent advancements in underground construction have led to the widespread utilization of pipe jacking. However, the engineering challenges posed by frozen ground in pipe jacking projects have not been extensively studied. This research aims to address the critical challenges linked to employing pipe jacking in frozen ground for underground construction. It is widely recognized that the accurate calculation of jacking thrust and mitigation of pipe–soil interaction plays a crucial role in determining the success or failure of pipe jacking operations. To explore these issues, this study conducted numerical simulations and comparative analyses, considering various factors such as soil properties, geometric dimensions, and burial depth, to assess their influence on jacking thrust. Additionally, the study also examines the freeze–thaw effect on concrete pipes and the injected lubricant. The results indicate that the numerical model, which considers the temperature effects and static friction instead of sliding friction, provides a more reliable estimation of jacking thrust in frozen ground compared to traditional theoretical models. Furthermore, the freezing point depression method was successfully employed in the development of an anti-freezing lubricant, which can effectively reduce pipe–soil interaction even at extremely low temperatures of up to −10 °C.

1. Introduction

Underground engineering, including for the subway network, pipeline network, and integrated pipe gallery, is rapidly advancing. Pipe jacking is a trenchless method for crossing roads, railways, rivers, and other complex geological conditions, minimizing open excavation or traffic disruption [1,2,3]. Pipe jacking is generally successful, but faces challenges in soft soil with low standard penetration, fine sand with rich groundwater, soil layers with strong spheroid-weathered granite, and strata with soft and hard interaction. As a common technical means of water sealing and strengthening in water-rich and weak strata and an environmentally friendly technology, the artificial freezing method does not affect the water table and water quality, and it is very suitable for water-rich and weak strata strengthening and sealing [4]. The artificial freezing method is commonly used in mine construction, with limited research on its application in urban underground engineering [5,6]. Currently, pipe jacking is seldom used in frozen soil, and research results are scarce. Hence, this study aims to delve into the pertinent matters when deploying pipe jacking technology in frozen ground-based underground construction.

Jacking force is crucial for designing the pipe wall thickness, intermediate jacking station location, frame selection, and lubrication requirements [7,8,9]. On the one hand, accurate calculation of the jacking thrust will not cause stress concentration, leading to structural failure of the pipeline. On the other hand, it will not cause high cost or even pipe jacking failure due to underestimation [10,11]. Frictional resistance between the pipe and the surrounding soil constitutes the primary component of the jacking load [2]. Pipe jacking is an inherently complex process, making it challenging to theoretically determine the required thrust force. The interaction between the pipe and soil is significantly influenced by excavation factors, including over-excavation, lubrication, pipe misalignment, stoppages, ground stability, and soil quality [12]. Determining jacking thrust requires calculating both the face resistance and the frictional forces acting along the pipeline. The thrust needed to counteract frictional forces is constrained by the ultimate strength of the concrete pipe, which in turn governs the jacking distance [2]. Meanwhile, some methods like machine learning or deep learning, which can analyze large datasets collected from past projects, are used to predict the jacking force [13,14,15]. But till now, few research studies have concentrated on the pipe thrust predictive model, including the interaction of the pipe and frozen soil.

Using lubricant effectively reduces frictional resistance around the pipe, allowing for increased maximum jacking length [2,7]. And, the distribution and penetration of lubricant around the pipe in the overcut area affect the contact properties between the pipe and frozen soil. At the same time, because of the frozen temperature, the ordinary drag-reducing mud is very likely to freeze, which may result in a reduction in the lubrication effect, and the pipe section can even become “locked” within the frozen ground and cannot successfully conduct pipe jacking [16,17]. Due to the unique conditions of pipe jacking in frozen soil, anti-freeze drag-reducing mud must provide anti-friction and support effects, along with good flow under low-temperature conditions. Recent advancements have introduced effective methods for stabilizing unstable strata by using tail-void mud slurry to exclude water from excavations. This system meets modern environmental standards, working well in normal-temperature strata of pipe jacking engineering. However, there have been no reports on anti-freeze resistance-reducing mud that can be directly used in pipe jacking within frozen soil. Only a small amount of anti-freeze drilling mud has been used in exploration projects, which is essentially different from the lubricant used in pipe jacking. Thus, it is necessary to develop anti-freezing lubricants that perform effectively at subzero temperatures to ensure successful pipe jacking in frozen soil.

2. Numerical Model of Jacking Thrust

2.1. Frozen Soil

Frozen soil represents a natural system with multiple components and phases, and its mechanical properties undergo influences from various factors. These factors include temperature, dry density, water content [18], the size and distribution of mineral particles, strain rate, sample size or shape, loading conditions [19], and so on. Over the past few decades, numerous experiments have been conducted by researchers to explore the deformation mechanism and mechanical characteristics of frozen soil [20,21].

In the Cold Regions Research and Engineering Laboratory, David and Zhu [22] executed a series of Uniaxial Compressive Strength (UCS) tests on remolded and saturated Fairbanks frozen silt. Their results led to the conclusion that the initial tangent modulus is primarily determined by temperature, as shown in Figure 1. Therefore, the expression for the initial tangent modulus can be formulated as follows:

$$E_i={\displaystyle \frac{2}{5}}{\left(\theta /{\theta }_0\right)}^{0.636}$$

(1)

where E_i represents the initial tangent modulus, θ is the temperature, and θ₀ is −1 °C. For the purposes of this study, frozen Fairbanks silt is selected as the target material for the subsequent numerical analysis, as its material properties are relatively well understood.

2.2. Pipe–Soil Interaction

As previously mentioned, the primary components of thrust force encompass face resistance and frictional force between the pipe and the surrounding soil, with frictional force being dominant. Consequently, the magnitude of the jacking force is contingent upon the interaction between the pipeline and frozen ground. The friction coefficient μ, which is a function of the friction angle δ between the concrete pipe and the frozen soil, plays an essential role in governing these interactions. The friction angle is typically assumed to range between an upper limit equal to the internal friction angle of the soil (φ) and a lower limit, which usually varies from φ/2 to φ/3, depending on factors like the roughness of the interface and the degree of movement [1].

When water freezes, the pipe interacts not only with soil particles but also with ice crystals. The ice acts as a binding agent, combining neighboring soil particles or rock fragments, thereby increasing their strength and preventing water seepage [23]. This suggests that the presence of ice crystals changes the way the pipe interacts with the surrounding ground [24]. As the water between the pipe and soil freezes at subzero temperatures, the bond force intensifies, which may block the pipe and make it challenging to accurately determine the real contact conditions. Meanwhile, the application of anti-freezing lubricant modifies the interaction between the concrete pipe and frozen ground. Figure 2 provides a schematic representation of the interaction between the pipe and frozen soil.

2.3. Interface Contact Types

The jacking force’s frictional component consists of two main factors: friction between the pipe and the soil, and friction between the pipe and the lubricant. In order to validate the thrust force, the influences of over-excavation and lubrication were incorporated by adjusting the contact range and friction coefficients between the pipe, soil, and lubricant using Fast Lagrangian Analysis of Continua in 3 Dimensions (FLAC3D). In FLAC3D, an interface is viewed as a collection of triangular elements, each of which is defined by three nodes. It is recommended to use the lowest stiffness consistent with small interface deformations, set to 10 times the equivalent stiffness of the stiffest region (Itasca, 2012). After using lubricant, the bottom and pipe are no longer in direct contact, but in contact with the lubricant. To assess the cohesion and friction angle of the lubricant interface under various normal pressures, Khazaei et al. [2,21] conducted direct shear tests on the commercially available lubricant DKIα, a blend of two components: agent A (Na₂SiO₃) and agent B (KHCO₃). Their findings indicate that the cohesion and friction angle at the lubricant interface are 2.66 kPa and 9.89°, respectively. These two interface properties were selected as lubrication regions. Conversely, other contact characteristics are modified based on varying friction coefficients between the concrete pipe and frozen soil, which are influenced by ground temperature, as demonstrated by Liu’s [25] results. The behavior of the contact surface includes tangential behavior and normal behavior [26]. The contact pressures and stresses on the outer surface of the pipe are directly generated by the weight of the surrounding soil. The pipe jacking machine has a super cutting and digging imitation milling function, which helps reduce the contact area and friction between the ground and the pipe. Based on previous studies [26,27], it is considered that the contact conditions within the range of 1/2 and 1/3 are more distinguishing and representative. The contact range parameters were defined as full, half, and one-third of the pipe surface in contact with the surrounding ground, as illustrated in Figure 3a–c.

2.4. Numerical Simulation

A thermal–mechanical model was developed to account for ground temperature, with the mechanical simulation being carried out based on the results from the thermal model. Initially, thermal simulation is carried out, followed by excavation on the same model based on the temperature distribution. The simulation process involves several key steps: First, a thermal simulation is conducted with the bottom boundary at 10 °C and surface boundaries between −5 °C and −20 °C. Then, the initial stress state is established, and soil properties are adjusted using the embedded FISH function to reflect the relationship between the modulus and temperature. The contact interface between the concrete pipe and the frozen ground is defined based on three typical contact types.

Table 1. Numerical parameters.

Geometric parameters

Jacking length (L): 20, 40, 60, 80 m; Diameter (D): 1, 2, 3 and 4 m; Cover depth (C): 4, 6, 8 and 10 m.

Geological properties

Friction angle (φ): 20, 30, 40 and 50°; Cohesion (c): 10, 30, 50 and 70 kPa.

Interaction properties

Friction coefficient (f): f (−5 °C), f (−10 °C), f (−15 °C) and f (−20 °C); Contact range: full, half, and one-third contact.

Concrete pipe

Concrete pipe: C40, Ec = 3.25e4 MPa, μc = 0.2.

lists the parameters that affect the numerical analysis. The jacking distance ranges from 20 to 80 m, which is used to estimate the jacking force as it relates to the distance. A considerable jacking thrust accompanies long-distance pipe jacking, imposing limitations on the one-time jacking distance, necessitating intermediate jacking stations. Consequently, the ultimate strength of the concrete pipe is factored in to compute the one-time jacking distance, thereby determining the requirements for intermediate jacking stations and lubricants. This study selects the C40 concrete pipe in the numerical simulation [28].

2.5. Numerical Model

As previously stated, the numerical model functions as a coupled thermal–mechanical system. To mitigate boundary effects, the model dimensions are established at a width of 40 m and a height of 30 m. Boundary conditions include hinge constraints on the bottom face, roller supports on vertical faces, and a free boundary for the ground surface. The model incorporates an interface to simulate interactions between the concrete pipe, frozen ground, and lubricant. Figure 4 illustrates the thermal and mechanical models.

2.6. Results Analysis

2.6.1. Effect of Jacking Length

In Figure 5, the thrust force outcomes are presented, illustrating variations in jacking distance from 20 to 80 m across the entire contact range. Graphs (a)–(e) delineate the impact of diverse factors, namely ground temperature, cohesion, friction angle of frozen soil, diameter, and burial depth, respectively. Evidently, the jacking force exhibits an almost linear escalation with the increase in jacking distance, irrespective of the prevailing contact conditions.

2.6.2. Effect of Temperature

Figure 6a–c depict the correlation between ground temperature and jacking force under three contact ranges: full, half, and one-third contact, respectively. Observably, the thrust force diminishes with rising ground temperature. This results from pore water freezing at lower temperatures, strengthening the bond between the pipe and ground. Furthermore, the jacking force for full contact surpasses that of half and one-third contact, affirming the significant reduction in thrust force with lubricant application. In Figure 7a, the alteration in ground temperature is shown to affect the maximum jacking length. Notably, lubricant injection substantially extends the maximum jacking length. The maximum jacking length decreases with lower temperatures, especially without lubricant.

2.6.3. Effect of Soil Properties

Figure 6d–f illustrate the correlation between jacking force and the cohesion of frozen soil under three contact ranges at T = 5 °C, φ = 20°, D = 2 m, and C = 6 m. The thrust force marginally decreases with increasing cohesion; however, the rate of increase diminishes between 30 and 70 kPa. Higher cohesion stabilizes the ground, reducing normal stress on the pipe surface. Figure 7b shows the relationship between maximum jacking length and cohesion of frozen ground. The maximum jacking length increases slightly with cohesion, but the effect is mild. Lubricant injection increases the maximum jacking length from 50 to 80 m.

Figure 6g–i depict the impact of the friction angle of the frozen ground on jacking force and maximum jacking length at T = 5 °C, c = 10 kPa, D = 2 m, and C = 6 m. The thrust force decreases as the friction angle increases. Figure 7c presents the effect of friction angle change on maximum jacking length. The maximum jacking length rises with an increasing friction angle, particularly from 20 to 30°, but the increment diminishes from 30 to 50°. Likewise, lubricant injection increases the maximum jacking length by about 30 m.

2.6.4. Effect of Pipe Diameter

The impact of pipe size on thrust force and maximum jacking length is investigated by varying it from 1 to 4 m. Figure 6j–l demonstrate the substantial influence of the pipe diameter on both the thrust force and maximum jacking length. The results reveal a linear increase in thrust force with tunnel diameter, indicating that larger diameters correspond to higher thrust forces. This is attributed to the larger diameter enhancing the contact area between the pipe and soil. Simultaneously, the maximum jacking length decreases rapidly as the diameter of pipe increases. For long, large-diameter pipelines, proper placement of intermediate jacking stations and even lubricant distribution are crucial for achieving the desired results, as shown in Figure 7d.

2.6.5. Effect of Burial Depth

Figure 6m–o depict the correlation between jacking force and jacking distance at various burial depths under three contact conditions. The results reveal a rapid increase in thrust force with growing jacking distance. Moreover, the thrust force rises with the increase in burial depth. Figure 7e illustrates the maximum jacking length’s variation with changing burial depth. It is evident that the maximum jacking length decreases swiftly with an increase in burial depth, particularly from 4 to 6 m and from 6 to 10 m. This reduction occurs linearly with the ascending burial depth. Furthermore, when comparing different contacts, the one-third-contact condition exhibits the minimum jacking force and maximum jacking length.

2.7. Comparative Analysis with Theoretical Solutions

Common equations for predicting thrust in pipe jacking through unfrozen soil have been derived from the literature, listed in

Table 2. Review of the jacking force models [31].

Model	$\mathbf{Frictional} \mathbf{Component}{\mathit{F}}_{\mathit{f}\mathit{r}\mathit{i}\mathit{c}}$	Symbols and Notes
JSTT	$F_{fric}={\mu }^{’}L\beta \left\{B_{c}q+W+B_c{C}^{’}\right\}$	$\beta$—jacking force reduction factor (0.45 for sand soil); $B_c$—circumference of the pipe; $q$—normal force from Terzaghi model; $W$—the weight of pipe; $C^{’}$—adhesion of pipe and soil; N—the value of SPT $\mu =tan\left(\phi /2\right)$; $L$—pipeline or tunnel length.
CSTT	$F_{fric}=BcL{f}_k$	$b_a$ = machine external width; $d_a$ = machine external height; $f_k$—circumferential frictional resistance $K_2$—the thrust coefficient of soil acting on the pipe with a suggested value of 0.3, $K$—lateral earth pressure coefficient; ${\sigma }_{EV}={\displaystyle \frac{b(\gamma -2C/b)}{2Ktan\phi }}\left(1-e^{-2K(h/b)tan\phi }\right)$ $D_e$—the outer pipe diameter $\delta$—the angle of friction between the pipe and the soil; ${\sigma }_v={\displaystyle \frac{\gamma b}{Ktan\phi }}\left(1-e^{-K(h/b)tan\phi }\right)$; $b={\displaystyle \frac{tan(\pi /4-\phi /2)D_e}{2}}+{\displaystyle \frac{D_e}{2sin(\pi /4+\phi /2)}}$; ${\sigma }_h=k({\sigma }_v+0.5\gamma D_e)$
Pellet-Beaucour and Kastner	$F=\mu L{D}_e{\displaystyle \frac{\pi }{2}}\left[\left({\sigma }_{EV}+{\displaystyle \frac{\gamma D_e}{2}}\right)+K_2\left({\sigma }_{EV}+{\displaystyle \frac{\gamma D_e}{2}}\right)\right]$
PJA	$F=\mu L{D}_e{\displaystyle \frac{\pi }{2}}\left({\sigma }_v+{\sigma }_h\right)tan\delta$

. Despite the differences between unfrozen and frozen soil, the calculation model for thrust force through frozen ground can be estimated by using calculation equations developed for unfrozen soil, considering stress and contact state. The icy crystals in frozen soil alter the pipe–soil interaction, causing calculation errors when common models are used for frozen soil. Therefore, in order to illustrate the differences, the friction force deduced from the numerical simulation is compared with the calculation models frequently used by the Japan Society for Trenchless Technology (JSTT) [29], Pipe Jacking Association (PJA) [30], China Society for Trenchless Technology (CSTT), and Pellet-Beaucour and Kastner [1].

Figure 8 presents a comparative analysis of numerical and theoretical results, both with and without lubricant. In both lubricant and non-lubricant scenarios, the thrust force from numerical simulation exceeds that of theoretical solutions. This discrepancy may arise from geological conditions, specifically the freezing of water content, leading to a significant increase in cohesion and enhancing the bonding strength between the concrete pipe and frozen ground. It is important to note that the numerical simulation considered maximum static friction, unlike the sliding friction commonly used in theoretical equations. In non-lubricant pipe jacking, the CSTT’s theoretical model provides the most approximate estimation compared to numerical results. The formulas from JSTT and PJA make a relatively large estimation error. Similarly, within lubricated conditions, the numerical results exceed the jacking force calculated from theoretical equations with a big gap.

Therefore, caution should be exercised in designing and estimating jacking force through frozen ground using theoretical equations derived from unfrozen ground. The reason is that the commonly used theoretical models always underestimate the thrust force. The reason may be that common equations lack consideration of temperature effects and rely on sliding friction instead of static friction. The jacking force in conventional ground is smaller than that in frozen soil with reasonable estimation. This is because the existence of icy crystals in frozen soil enhances the cohesive action between the pipe and the surrounding soil. Therefore, the proposed numerical model in this paper offers a more reasonable estimation of thrust force for pipe jacking through frozen ground than the widely used theoretical models.

2.8. Freeze–Thaw Effect

Freezing–thawing cycles are a common construction phenomenon in frozen ground, leading to potential collapse and damage [32,33,34]. As previously mentioned, the insertion of concrete pipes and injected lubricant disrupts the thermal status of the surrounding frozen ground, resulting in a reduction in the mechanical properties of frozen soil. Therefore, a numerical model is employed to simulate the temperature field, considering the thawing effect of the pipe and lubricant. The temperature of concrete pipes and injected lubricant is set to 10 °C [35,36]. Figure 9 illustrates the temperature distribution without and with thawing effects. The temperature distribution changes from horizontal to non-horizontal, creating a melted area around the excavation, which decreases the mechanical properties of the surrounding frozen ground.

The results in Figure 10a–c show that jacking force is lower when thawing effects are considered, regardless of the contact conditions between the pipe and soil. The melting of frozen soil weakens the bond between the pipe and soil, reducing the jacking force. As the surrounding ground thaws, temperature changes have a smaller effect on the thrust force. Simultaneously, Figure 10d–f demonstrates that the maximum jacking length increases due to the melting of frozen ground. Ground temperature has minimal influence on the maximum jacking length when thawing effects are present.

3. Anti-Freezing Lubricant Development

3.1. Freezing Point Depression Method

As is commonly understood, typical lubricants can freeze under negative temperature conditions. Hence, there is a need to explore the lubricant’s performance under freezing temperatures and innovate an anti-freezing lubricant. This innovation is crucial to guarantee the success of pipe jacking in frozen ground. To avert freezing during the jacking process, the freezing point depression method was employed to reduce the freezing point of lubricant. Figure 11 illustrates the vapor–pressure curve for the solid with a dissolved solute at the freezing point of a solution, wherein the freezing point is lowered. The freezing point depression, ΔT_f, is a colligative property of a solution equal to the freezing point of the pure solvent minus the freezing point of the solution [37,38].

$$∆T_f=i{K}_f{c}_m$$

(2)

where ΔT_f—change in temperature in °C; K_f—freezing point depression constant for water, K_{f water} = 1.86 °C·kg/mol; i—Van’t Hoff factor; i = 2 for NaCl; i = 1 for sugar; and cm—molarity, mol/L. Utilizing the characteristics of the chosen lubricant materials, sodium chloride (NaCl) is selected as the solute in the freezing point depression method to decrease the lubricant’s freezing point. Figure 12 illustrates the correlation between freezing temperature and NaCl molarity.

3.2. Experiment Preparation

3.2.1. Lubricant Components

To formulate the anti-freezing lubricant, conventional slurry additives from Japan are employed to assess fluidity behaviors, encompassing gelling time and viscosity under low temperatures. The lubricant comprises two components: Component A, mainly consisting of silicic acid, sodium salt, and sodium silicate, and Component B, primarily composed of sodium bicarbonate, potassium bicarbonate, and a mixture of acrylic acid and light metal salt. The preparation of solutions A and B adheres to the specified concentrations outlined in

Table 3. Preparation of solutions A and B.

Composition	Water/g	Material A/mL or B/g
Solution A	82	18 mL A
Solution B	95.5	10 g B

. In total, nine temperature levels are designated as −20 °C, −15 °C, −10 °C, −5 °C, 0 °C, 5 °C, 10 °C, 15 °C, and 20 °C in the experiments. Triplicates of mixed solutions are prepared for each temperature, and the final results represent the average gelling time.

3.2.2. Experimental Equipment

The temperature and humidity chamber (CSL-1) serves as the temperature controller, operating within a temperature range of −20 to 100 °C and a humidity range of 20% to 95% RH. To assess the viscosity of the enhanced lubricant, the Brookfield AMETEK DV1 (Middleboro, MA, USA) viscometer is employed. This viscometer offers simultaneous measurements of viscosity and temperature, featuring an optional temperature probe and a selection of 18 speeds. For clarity in result interpretation, the HB-05 spindle is selected, operating at a rotational speed of 12 RPM. Figure 13 illustrates the experimental setup.

3.3. Experimental Results

3.3.1. Gelling Time of Anti-Freezing Lubricant

Figure 14 depicts the correlation between temperature and gelling time. As temperature decreases, gelling time shortens due to the slowed chemical reaction speed of solutes in the mixture solution. Additionally, elevating the sodium chloride molarity decreases the solution’s gelling time, attributed to sodium chloride’s accelerating effect on the chemical reaction.

Throughout the experiments, when the lubricant mixture lacked additional sodium chloride, it froze. Thus, it is both reasonable and imperative to include sodium chloride in the lubricant to prevent the jacking pipe from getting stuck in freezing temperatures. In comparison to scenarios without sodium chloride, the gelling time for the original formulations of solutions A and B increases more rapidly as temperature decreases than formulations with sodium chloride. The variation in gelling time with different amounts of sodium chloride is relatively minor across various temperatures. Even at temperatures between −10 °C and −5 °C, the gelling time can reach approximately 30 s, nearly matching the original slurry recipe’s gelling time under normal temperatures. Taking the gelling time of the original recipe at room temperature as the acceptable threshold for the developed anti-freezing lubricant, it is evident that the developed lubricant performs effectively under subzero ground temperatures, as indicated within the red rectangle in Figure 14. Additionally, the sodium chloride’s molarity should not exceed 2.69 mol/Kg. Beyond this concentration, gelation is scarce. These findings demonstrate that an anti-freezing lubricant can be achieved by elevating the sodium chloride molarity in the lubricant solution.

3.3.2. Viscosity of Anti-Freezing Lubricant

The viscosity, a property resisting relative motion between fluid surfaces moving at different velocities, measures a fluid’s resistance to gradual deformation under shear or tensile stress. In pipe jacking lubricants, this characteristic is exploited to diminish frictional resistance and fortify the surrounding ground with residual strength. Figure 15 illustrates viscosity changes concerning gelling time. These figures depict viscosity increasing with gelling time, with a swifter rise at higher temperatures due to the quicker chemical reactions in the lubricant solution. Lower temperatures decelerate chemical reactions. Comparative analysis reveals that sodium chloride addition expedites gelation, with higher concentrations accelerating the chemical reaction within the lubricant. Simultaneously, the maximum viscosity diminishes as the temperature drops.

3.3.3. Gelling States

The addition of sodium salt to conventional lubricants aims to decrease the freezing point in mixed solutions, thereby preventing adhesion during frozen soil pipe jacking operations. As previously discussed, incorporating sodium salt reduces the freezing point of lubricant, resulting in decreased friction and drag during pipe jacking through frozen soil. The introduction of sodium salt disrupts the original gelling chemical reaction in solutions A and B, leading to a modification in the final cementation form. Figure 16 illustrates how different concentrations of sodium chloride affect the state of cementation. Higher concentrations of sodium chloride led to weaker cementation strength, subsequently reducing viscosity and weakening the anti-freezing drag-reducing lubricant’s ability to support the surrounding rock mass within the overcutting space. However, it does not impact its effectiveness as a drag-reducing agent for lubrication purposes. To enhance the support provided by the lubricant, grouting pressure can be adjusted accordingly. In future studies, we will explore how adding mineral components, like bentonite, affects the supporting effect provided by lubrication mud.

4. Conclusions

Addressing challenges in pipe jacking within frozen ground, this paper explores the pipe–soil contact mechanism, thrust force estimation, and development of anti-freezing lubricant. It examines factors influencing jacking force, including jacking length, temperature, property of frozen soil, tunnel size, burial depth, and lubricant distribution around the pipe, providing comprehensive insights, with the purpose of verifying the prediction accuracy of the commonly used predictive model of thrust force when pipe jacking passes through the frozen ground. In order to guarantee the successful operation of pipe jacking in frozen soil, an anti-freezing lubricant is developed by using the freezing point depression approach to ensure successful pipe jacking in frozen soil construction. The conclusions drawn in this article are as follows:

• The numerical method for predicting jacking force in this paper is more reliable than conventional models. The freezing of pore water strengthens the bond between the pipe and frozen ground. In addition, the new model uses maximum static friction, unlike the sliding friction used in theoretical models, leading to higher jacking force estimates.
• The insertion of concrete pipe and injected lubricant disrupts the thermal status of the surrounding frozen ground, resulting in a reduction in the mechanical properties of frozen soil. Furthermore, as the frozen ground around the pipeline thaws, the changing ground temperature has a diminished impact on the jacking force.
• The freezing point depression method is applicable to develop an anti-freezing lubricant for pipe jacking in frozen ground, which can ensure the successful execution of pipe jacking even in freezing temperatures as low as −10 °C.
• The introduction of sodium salt disrupts the original gelling chemical reaction in solutions, leading to a modification in the final cementation form, which subsequently reduces viscosity and weakens the supporting effect within the overcutting space.

In the future, we are trying to build the calculation model of jacking force taking into consideration temperature, especially for water-rich and weak strata. On the other hand, long-term experiments under real-site conditions will be performed to verify the performance of the anti-frost lubricant.